
For years, you have painstakingly built higher and higher rankings in Google. Your content answers exactly what users are searching for. Then your traffic suddenly drops, for some by as much as 30% a year or more. Welcome to the world of zero-click search.
This challenge is already here. It is the reality out there, and the pace of change keeps accelerating. Anyone with a digital platform where traffic is business-critical faces a strategic choice.
Zero-click search occurs when a potential customer gets the answer directly in the search results, without clicking through to a website. This can happen through Google’s AI Overviews, featured snippets, knowledge panels or AI chat tools such as Claude, ChatGPT and Perplexity.
The phenomenon is old. What is new is the speed over the past year. These tools are being rolled out at a high pace in ever more markets. In Norway we see the same pattern as internationally, with one important difference: Norwegian businesses are generally slower to react to search trends than companies in the UK and the US.
AI services are changing the rules of the game too. ChatGPT has over 1.2 billion weekly users, OpenAI said at its DevDay developer conference on September 29, 2026. Perplexity handled 780 million searches in May 2025 alone, with more than 20 percent growth per month. At the same time, AI tools accounted for under 2 percent of desktop visits in March 2026, according to Datos and SparkToro. Search is shifting, but gradually.
These platforms are designed first and foremost to give users answers, and they send little traffic on to your website. Even so, the data shows that users want to validate an AI recommendation before they move on.
This means the customer journey increasingly starts in the AI interface, and the journey continues from there: it actually becomes longer and more complex.. The user asks AI a question, gets a curated answer with recommended suppliers and solutions, and then moves on to validate the recommendations in Google and on selected websites.
Instead of all traffic going straight from Google to individual pages, we see brand searches pick up after an AI answer has been delivered. The user typically searches further on variations such as “your brand + reviews,” “your brand + price” or “your brand compared with competitor” to double-check what the AI has suggested. The starting point of the customer journey has changed, and the journey itself has gained an extra layer of validation.
Read more

New digital booking platform gave record sales for DE BERGENSKE
For several customers, we therefore see a combination of fewer generic, non-branded clicks and growth in brand searches and direct traffic when they are mentioned often in AI answers. The strategic task is to make sure you are both recommended by AI and easy to find and choose when the user moves on to Google to verify the recommendation.
Google, of course, is well under way turning this relatively new format into business models. The share of AI Overview searches that also include Google Ads grew from under 3% in January 2025 to around 40% in November, and Google is now testing ads directly inside the AI answer itself.
AI Overviews are now moving fast from covering only informational searches to capturing purchase intent and navigational searches. That means the zero-click tsunami is now just as relevant for those who sell products as for those who produce and publish content.
Everything can be turned into an opportunity, these changes included. But it is crucial that you understand what is happening, why it is happening and how you can use it to your advantage.
When AI visitors actually click through to your website, they are much further along in the decision process than a typical organic searcher. They have already done their research and are ready to decide. This shows up in conversion rates that are documented five times higher than for organic Google traffic.
The second effect is at least as important: repeated exposure. When AI consistently mentions your brand name in authoritative contexts, it builds trust and recognition that trigger searches and clicks in other channels. Brands that are cited in AI Overviews actually get more clicks, not fewer. The mechanism is logical: even if the user does not click right there and then, they remember your name the next time they meet it.
It is important to note the following. Google’s total search volume is growing, and the growth is still significant. Even though the share of zero-click searches rises quickly, the absolute number of clicks has actually increased. There are as many clicks to share as before, but the fight for them has become harder.

Based on the work we do with Norwegian clients every day, three mistakes keep recurring, regardless of industry and size.
No Wikipedia page, no mentions in trade media, no activity in professional forums or on Reddit. AI systems judge credibility by triangulating presence across platforms. A business that exists only on its own pages comes across as an unknown player, no matter how good the content is.
The dashboard shows Google Analytics, and all that counts is sessions and pageviews. Brand searches, direct traffic and AI citations go unmeasured, which means half the picture is missing. You can lose visibility for weeks without anyone noticing.
When users first get recommendations in AI and then go to Google to validate, the value shifts from anonymous clicks to clear intent signals. If branded queries, direct visits and conversion on pages that AI often mentions are rising, it is a sign that your visibility is working, even if total traffic does not necessarily grow.
Without these data points, you can lose visibility in AI and on brand while the session curves look “OK.” You react only when competitors have already taken the positions in both the AI interface and the Google results.
Content optimized only for keywords, meta titles and descriptions hits the mark less and less often. Google now rewrites titles at scale when they seem too long, keyword-stuffed or imprecise: in Q1 2025, around 70–76% of all title tags were fully or partially changed before display in the search results.
The title the user sees can therefore differ from the one you wrote. The algorithms prioritize short, clear titles that mirror the content and the search intent, even if that means dropping the focus keyword or the brand name. Meta descriptions are also increasingly generated directly from the page content, based on what Google believes answers the user’s query best.
At the same time, AI tools pull the answer from the content itself rather than from the meta text. Both Google and generative search engines prefer clear, structured answers that explicitly address the questions people actually ask, over texts that simply repeat keywords in the title and intro.
You need to write content that holds up even when Google rewrites your title, and when AI lifts out individual paragraphs as the “answer key.” You do that by placing the most important answer early, using clear subheadings and FAQ structure, and phrasing yourself as if you are answering a specific question.
In a zero-click world, the methodology is in many ways turned on its head. Here is how.
GEO (Generative Engine Optimization) is the framework for achieving visibility in AI-generated answers. Research from Princeton shows that GEO techniques can increase AI visibility by up to 40%. Here is how it works:
AI also reads what others write about your brand, in addition to your website. The more places your name appears with consistent and credible information, the more authority both search engines and language models grant you as an entity.
Recent studies show that nofollow links correlate almost as strongly with visibility in AI answers as regular links. The models care more about context and the authority of the source than about the attribute itself.
A brand that in practice exists only on its own website will therefore lose to players that are regularly mentioned and linked to from relevant, credible sources, whether the links are follow or nofollow. The task is to build a visible, recognizable footprint on the platforms AI actually reads and learns from.
It is still smart to get to the point quickly, but AI and Google do not depend on your answer sitting in the first 40–60 words of every paragraph. They use the whole page, internal link structure and structured data to find the most precise and credible answers, even when those sit further down the page.
That is why your job is less about “squeezing everything into the intro” and more about making your message explicit and machine-readable. With schema markup (for example Article, FAQPage, HowTo and Organization) you can mark which parts of the content are questions and answers, which paragraphs describe a process, and who is behind the content.
Properly structured data works as a map for both Google and AI models: you clearly point to the main answer, which definitions matter and which FAQs should be lifted forward. Pages that combine well-written content with robust schema markup are used more often in AI overviews and other generative answers than pages that are only “good” at traditional on-page SEO and well-written content.
Headings, tables, numbered lists, comparisons and definition boxes are 28–40% more likely to be cited than running prose without clear structure.
Include reliable statistics, sources and data.
Content with numbers and source references gives up to 28% higher AI visibility. Original data and your own surveys rank even higher. And make sure every fact you pass on is correct, because content with factual errors is penalized.
Media outlets and online newspapers often block AI bots to protect their content. For the vast majority of other businesses, blocking GPTBot, ClaudeBot or PerplexityBot is a direct strategic mistake. Without access, the AI services cannot cite you, and you lose visibility in a channel that is growing.

Traffic has long been the first number everyone looks at. That is understandable: it is easy to measure, and it feels concrete. But in a world where most searches never send anyone to your website, traffic alone gives an incomplete picture of how visible you really are.
Traffic still matters. You also need more measurement points.
Start by mapping how often your brand shows up in AI-generated answers, in ChatGPT, Perplexity, Claude and Google AI. Is it you who gets cited, or a competitor? Is the number of searches where people actively look for your brand growing? Is direct traffic increasing over time? These are signals that zero-click exposure is actually building awareness and trust, even when nobody clicks right there and then.
Finally, look at what the AI traffic is actually worth. Conversion rate and revenue per visitor from AI channels tell you more about the value of your visibility than a traffic figure alone ever can.
Only 22% of marketers track AI visibility systematically today. That is a head start that is relatively easy to take, if you begin now. 
Zero-click search is here to stay, and it pays to adapt now. The shift is already under way, and the pace is increasing. Those who react first win.
The good news: visitors who click through from AI sources are far closer to a purchase decision than a typical organic search visit.
It is tempting to treat zero-click as a technical SEO problem that can be solved with schema markup and FAQ sections. It is more than that. It is a question of who AI perceives as the authoritatively correct source in your field. That position is built over time, with consistent, high-quality content and no quick hacks.
Read more in the glossary about AI visibility, GEO and AEO.


Read more

700 meters inside the mountain: Lefdal Mine Data Centers

XXL: Black Week – Bright Moments